JPH065954B2 - Interframe and intraframe adaptive predictive coding device for moving image signal - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to digital transmission of image signals, and more particularly to a highly efficient band compression transmission technique based on an adaptive predictive coding system for moving image signals.

(Prior art and its problems) In predictive coding of a moving image signal, inter-frame prediction is generally suitable for a still image or a quasi-still image including slow motion, and intra-frame prediction is applied to an image including violent motion. It is known that it is suitable for. In addition, as a coding method that adaptively uses these two prediction methods, it is investigated whether inter-frame prediction or intra-frame prediction is more efficient for a coded pixel in the vicinity of the pixel to be predicted. , There is known a method of determining a prediction method for the pixel based on the result and a predetermined rule (Koga et al .: "Motion Compensation Frames of Video Conferencing Image Signals, Intraframe Adaptive Predictive Coding", Showa). Proceedings of the 59th IEICE General Conference, 1277). However, such a method estimates a prediction method for a pixel in the vicinity from the fact that which coding method is suitable for a neighboring pixel, and if the estimation is erroneous, the prediction efficiency may decrease. Inevitable. For example, FIG. 5 shows the case where the object A crosses the screen to the left, but for the region B newly appearing as the background of A, the intraframe prediction is generally more efficient. At this time, the inter-frame prediction is originally suitable for the boundary area between the still areas C and B, but in the conventional method, the intra-frame prediction is performed for the neighboring coded pixels (that is, the pixels in the area B). When the prediction is appropriate, the intra-frame prediction is still selected. Also, when the motion vector obtained in the Koga et al. Paper is a zero vector, there is a method of considering this as a still region and selecting interframe prediction, but this method is a locally decoded signal containing quantization noise. Is used to find the motion vector. Therefore, whether or not the motion vector is zero does not correspond to the actual motion accurately, and thus the improvement of the coding efficiency by the adaptive prediction cannot always be expected.

(Object of the Invention) An object of the present invention is to eliminate the above-mentioned drawbacks, to enable proper interframe prediction to be selected especially for a still region, and to provide a more efficient adaptive prediction coding system. .

(Structure of the Invention) According to the present invention, a moving image signal is supplied, means for intra-frame predicting the moving image signal and outputting an intra-frame predicting signal, and inter-frame predicting signal for inter-frame predicting the moving image signal. And a means for storing which of the intra-frame prediction signal and the inter-frame prediction is suitable for the already coded pixel, and the intra-frame prediction signal using the output of the storing means. , Between the frames of the moving image signal, and between frames of the moving image signal, comprising: selecting means for outputting any one of the inter-frame predicted signals as a predicted signal; and means for predictively encoding the moving image signal using the output of the selecting means. In the adaptive predictive coding device, regardless of the output of the means for separating the motion area and the still area of the input moving image signal and the storage means when the separation result is static. And selecting the inter-frame prediction signal by the selecting means, and outputting the selection signal to the selecting means based on the output of the storing means when the separation result is the moving region. An intraframe adaptive predictive coding device can be obtained between frames of a moving image signal.

(Detailed Description of Configuration) In the present invention, first, an input image signal before applying the predictive coding is separated into a motion area and a still area. As this separation method, for example, the screen is divided into blocks of a certain size, the frame difference value of each pixel in the block is added in the block, by the size comparison of this addition result and the predetermined threshold, A method of judging the motion of the block is used. That is, since the motion area is separated by using the input signal that does not include the quantization noise, the separation accuracy is high, and the influence of the error of the motion determination on the coding efficiency can be eliminated.

Description will be added with reference to FIG. When a certain pixel A is a pixel in a still area, the prediction method for this pixel selects an optimum interframe prediction method for the still area. In the case of pixel B in the motion area, intraframe adaptive prediction is performed during normal frames. For example, as described in the above-mentioned document, it is checked which of inter-frame prediction and intra-frame prediction is actually suitable for a plurality of pixels C, D, E, and F in the vicinity of a pixel to be predicted, and this combination and a predetermined number are determined. The prediction method of the pixel B is determined from the rule. That is, the present invention performs optimal inter-frame prediction for a still region defined by using an input signal and executes adaptive prediction for a motion region regardless of the determination result of the adaptive prediction method. .

Embodiments of the Invention Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 2 is an embodiment of an encoder using the present invention. Ten
The moving image signal input from 00 is supplied to the delay circuit 30 and the frame memory 10. Frame memory
10 delays the input signal by one frame and outputs the delayed signal to the subtractor 11. The subtractor 11 calculates the frame difference and outputs it to the intra-block adder 112. The in-block adder 12 is composed of an absolute value calculator, a necessary line memory, and an adder, calculates the sum of the frame difference absolute values of the pixels in the block, and outputs this to the comparator 13. The comparator 13 compares this with a predetermined threshold value and outputs the result as motion / static information, and outputs 1 to the code converter 16 and the determiner 20 if the result is large. The delay circuit 30 delays the input signal by the time required for the above-mentioned motion determination, and outputs it to the subtractor 14. The subtractor 14 calculates the difference between the input signal and the prediction signal supplied from the selector 17, and outputs the result to the quantizer 15 as a prediction error. The quantizer 15 quantizes the prediction error and outputs it to the code converter 16 and the adder 19.
The adder 19 adds the quantized prediction error and the prediction signal supplied via the line 1719, and outputs the result to the prediction circuit 18 and the optimum prediction determination circuit 22 as a local decoded signal. The prediction circuit 18 has, for example, a frame memory for inter-frame prediction, a line memory or a register for intra-frame prediction, and generates a plurality of prediction signals and supplies them to the selector 17 and the optimum prediction determination circuit 22. Optimal prediction decision circuit 22
Compares the locally decoded signal supplied from the adder 19 with a plurality of predicted signals, and if the prediction method that is closer to the locally decoded signal, that is, the optimum prediction method is inter-frame prediction, is 1 If it is 0, the memory circuit 21 is set to 0.
Output to. The storage circuit 21 stores the optimum prediction method of pixels in the vicinity of the pixel to be predicted as 1 and 0, and outputs this to the determiner 20. The determiner 20, the motion information supplied from the comparator 13, the inter-frame prediction when the static region is represented, otherwise supplied from the storage circuit 21, the frame for the neighboring encoded pixels. Information on which of the inter prediction and the intra prediction is the optimum prediction method is converted using a predetermined rule to determine the prediction method of the pixel and output to the selector 17. This judging device 20 is, for example, a ROM
Can be easily realized by using. That is, the motion information supplied from the line 1320 and the optimum prediction information for the neighboring coded pixels supplied from the line 2120 are used as address inputs, and the output is supplied as prediction method selection information to the selector 17 via the line 2017. To do. FIG. 4 shows an example of logic used in this ROM. M / S is, for example, motion information indicating that 1 is moving and 0 is stationary, and C, E, and F are optimum for neighboring pixels C, E, and F when the pixel B in FIG. 1 is a pixel to be encoded. In the prediction information, S represents the prediction method for pixel B (1 is interframe prediction). For example, M / S = 0, C = E = F = 0
In the case of, the intra-frame prediction is optimum for the neighboring pixels, but since it is a still region, S = 1, that is, inter-frame prediction is selected. The selector 17 selects one of the plurality of prediction signals supplied from the prediction circuit 18 according to the above-mentioned prediction method selection information S, and outputs it to the subtractor 14 and the adder 19. The code converter 16 converts the quantized prediction error signal and motion information into an appropriate code (for example, Huffman code),
Output to transmission line 2000.

FIG. 3 is an example of a decoder corresponding to the encoder described above.
The code inverse converter 23 inversely converts the encoded motion information and prediction error supplied from the line 2000, and the motion information is determined by the determiner 27.
The prediction error is output to the adder 24. The adder 24 is
Add the prediction signal and prediction error supplied from the selector 25,
The decoded signal is output to the output terminal 3000, the prediction circuit 26, and the optimum prediction determination circuit 29. Prediction circuit 26, optimal prediction determination circuit 2
9. The operations of the memory circuit 28, the determiner 27, and the selector 25 are the same as those of the encoder described above.

(Effects of the Invention) As described above, according to the present invention, it is possible to determine whether the input signal does not include quantization noise for the prediction error, and to implement an adaptive prediction method with a high probability of selecting an optimum prediction method.
Since the prediction adapted to the local property of the input signal is executed, the coding efficiency is improved, and the significance of supplying the present invention is great.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the operation of the present invention, FIG. 2 is a block diagram showing an embodiment of an encoder using the present invention, and FIG. 3 is a block diagram showing an embodiment of the same decoder. FIG. 4 is a diagram showing the logic of the FIG. 2 judging device 20 and the FIG. 3 judging device 27, and FIG. In the figure, 10 ... Frame memory 11, 14 ... Subtractor 12 ... In-block adder 13 ... Comparator, 15 ... Quantizer 16 ... Code converter, 17, 25 ... Selector 18, 26 ... Prediction circuit, 19, 24 ... adder 20, 29 ... decision device, 21, 28 ... storage circuit 22, 29 ... optimal prediction decision circuit 23 ... code converter, 30 ... delay circuit.

Claims (1)

[Claims] 1. A means for receiving a moving picture signal, predicting the moving picture signal within a frame and outputting an intra-frame prediction signal, and means for predicting the moving picture signal between frames and outputting an inter-frame prediction signal. Means for storing which of the intra-frame prediction signal and the inter-frame prediction is suitable for the already coded pixel, and the intra-frame prediction signal and the inter-frame prediction signal using the output of the storing means An intra-frame adaptive predictive coding apparatus for inter-frame of a moving picture signal, comprising: selecting means for outputting any one of the above as a prediction signal and means for predictively coding the moving picture signal using the output of the selecting means. , A means for separating the moving area and the still area of the input moving image signal, and if the separation result is static, the selecting means operates to select the And a means for outputting a selection signal to the selecting means based on the output of the storing means when the inter-frame prediction signal is selected and the separation result is a moving area. Interframe and intraframe adaptive predictive coding device.